Process parameter dependencies of continuous and pulsed laser modes on surface polishing of additive manufactured aluminium AlSi10Mg parts

2.1 Material and samples

The experimental investigations are executed on the aluminium alloy AlSi10Mg. Rectangular plates with the dimension of 100 mm length, 30 mm width and a material thickness of 2 mm were built up by selective laser melting (SLM) on a SLM280HL (SLM Solutions GmbH). The machine has a fabrication chamber size of 280×280×280 mm³ and a 400 W Yb-fibre laser. The aluminium powder particle distribution used (average powder grain diameter of 37 μm) can be taken from 3. The samples were built with recommended fabrication parameters, provided by SLM Solutions GmbH, with a slicing thickness of 50 μm. The layers are melted on the outer contour by 350 W laser power and a beam velocity of 600 mm/s. The core of the part is done with 350 W, 930 mm/s and a hatch distance of 170 μm.

After cutting the samples from the selective laser melting built platform (base plate) the initial surface of the 3D printed parts were analysed. Before laser polishing, the parts were cleaned by laser cleaning in order to ablate and reduce the oxide layer and the residual powder, which are left over from the 3D selective laser melting printing. The laser cleaning process were realized with a short pulse laser TruMark 5020 with a pump laser power of 20 W at a pulse duration of 70 ns and a pulse frequency of 65.5 kHz. The beam was guided over the surface with a track offset of 70 μm and a scanning velocity of 3000 mm/s.

2.2 Applied laser polishing setup

Laser polishing was carried out by means of a disk laser of type Trumpf TruDisk 4002 with a maximum output power of 4000 W. Two laser operation modes with pulsed wave (PW) and continuous wave (CW) were used to polish the aluminium surfaces, Figure 1. The laser beam was guided with a 200 μm gradient index fibre with a Numerical Aperture NA of 0.1 to a modified 5-axis Trumpf Laser Cell TLC 40, Figure 3.

The laser polishing is processed in a purified inert gas atmosphere of a sealed process chamber. An oxygen measuring instrument of type “PRO2 plus” controlled the residual oxygen concentration during the laser polishing process.

The beam deflection is carried out by means of a scanner optics of type SAO 1.06/1D, Figure 3. The scanner system enables a one dimensional (1D) oscillation of the laser beam with a maximum frequency of 300 Hz in combination with a maximum pendulum length of 70 mm. The average pendulum speed over the work piece surface is given by the pendulum frequency and the pendulum width x, equation 1.

(1)

In Figure 4, a schematic image of the pendulum movement, using the scanner optics, in conjunction with the machine axis is given.

The pendulum movement in y-direction is superimposed by the axis federate of the scanner head in x- direction. The resulting average beam velocity is calculated by the following equation 2:

(2)

The revealing area rate of laser polishing is calculated from the axis feed rate and the pendulum width , according to equation 3.

(3)

Due to the 1D pendulum movement, two turning points emerge, where the laser beam velocity in y-direction is momentarily zero, Figure 4 (two dotted black lines). With a constant laser power, the energy deposition on the turning points is significantly increased. As a result, the melt depth would be higher and process instabilities could appear. To prevent this, the pendulum axis is sectioned in the y-direction into 15 segments with adjustable laser power, Figure 5. In preliminary investigations, a progressive reduction of the adjusted nominal laser power P_{L, nom} in two steps of respectively 200 W at both turning points was found out as a convenient setup for the further investigations.

The resulting energy input per area, the energy density, also called fluency, depends on the laser power, the scanning velocity and the pulse and track overlap per area. For one polishing pass including the multiple energy input of the overlapping tracks or overlapping pulses is calculated by the following equation 4. For a areal treatment, energy density ED is given by the laser power , respectively the mean laser power at the pulsed laser mode, the pendulum width×and the axis velocity .

(4)

At the pulsed mode the resulting percentage pulse overlap PO_av is a function of the beam diameter at the surface d_L (z), the mean beam velocity and the pulse repetition rate according to equation 5.

(5)

2.3 Laser beam characterization

The laser beam was characterized by a caustic measurement device of the type primes focus monitor FM+ at a laser power of 500 W, Figure 6a. According to DIN EN ISO 11146-1 : 2020 the laser beam is measured over 21 measuring levels around the focus of the laser beam to get a complete beam caustic. The laser beam diameters are determined at 1/e² (86.5 %). The laser beam has a focus diameter d_f of 450 μm, a Rayleigh length z_R of 5.1 mm and a divergence angle Θ = 76.8 mrad. The M² of the measured beam amounts 30.2.

Additionally the laser beam was analysed at a typical laser power for laser polishing of 1700 W in steps of 2 mm from focal position to a defocus position of 20 mm. Figure 6a shows the laser beam diameter in relation to the distance in z-direction from the focal position.

The measured beam intensity profiles have a Gaussian intensity distribution. At focal position, the profile exhibits an intensity drop in the centre of the beam. The peak intensity is in the range of 2000 kW/cm². At 10 mm defocussing the peak intensity decreases to 650 kW/cm²–830 kW/cm². At 20 mm defocussing residual peak intensity amounts to less than 220 kW/cm², Figure 6b.

2.4 Measurement devices and evaluation methods

The initial and polished surfaces were investigated quantitatively and qualitatively. The quantitative analysis of the surfaces roughness, i. e. Ra and Rz, according to EN ISO 4288 : 1997, were realized tactilely by means of a perthometer of type MarSurf M400. According to standard ISO 4288 for the initial surface with Ra > 2 μm a cut-off wavelength of 2500 μm was used, whereas for the polished surfaces with Ra < 2 μm the cut-off wavelength is determined to 800 μm. The listed results of the polished surfaces are mean values based on five measurements. The positions of the single measurements were evenly distributed over the polishing field or test plates.

The surface is further analysed by a Fourier transformation of the measured tactile surface profiles. Hereto the tilt of the sample is deducted in the measurement data and levelled to zero. The result of the Fourier transformation is clustered into a logarithmic scale of the spatial wavelength. For each range of the spatial wavelength the arithmetic roughness Ra is calculated.

In this paper, polishing and measuring direction were varied. A schematic image of the layer wised selective laser melting samples with different polishing and measurement direction is depicted, Figure 7.

The initial surface topography and roughness were measured in the fabrication direction (FD) and perpendicular to fabrication direction of the selective laser melting process. In the fabrication direction, the roughness structures caused by the layer wise production technology are taken into account. Laser polishing was basically done with the fast laser beam deflection of the scanner optics in the fabrication direction and the superimposed axis movement perpendicular to the fabrication direction, Figure 7 (left side). Later, in the discussion of the results, one polishing movement perpendicular to the fabrication direction is defined as: 1×0, and one polishing pass in the fabrication direction as: 0x1. If laser polishing is repeated with two passes from one direction: 2×0 or a combination of one pass in fabrication direction and one pass across fabrication direction: 1×1;. two repeated passing's in both directions 2×2. After laser polishing, the surfaces were measured in the fabrication direction (FD) and perpendicular to fabrication direction. Qualitative evaluation was carried out with a 3D profilometer VR-3100 from Keyence. With this measurement, an optical evaluation of the 3D surface topography is done. The micro structure of the surface was analysed using scanning electron microscopy of type Carl Zeiss Sigma VP.

2.5 Constant process parameters during laser polishing

In this study, two different laser operation modes, continuous wave (CW) and pulsed wave (PW) laser radiation were applied. In both modes, a number of parameters were kept constant, Table 1. Several laser polishing parameters were applied on a test field matrix with field dimensions of 10×10 mm². The fields had a distance of 2 mm between each other and the edge distance was at least 3 mm. A latency time of 30 seconds between polishing of two fields was considered to prevent a significant rise in temperature of the aluminium substrate. The laser source was operating in pulsed mode with the laser specific maximum repetition rate of 1000 Hz and the shortest possible pulse duration of 0.3 ms (duty cycle 30 %) in order to maximize area rate. To achieve a smooth surface, the minimum pendulum frequency of the 1D scanner with 10 Hz was chosen to maximize the pulse overlap. The second investigation was conducted with continuous laser radiation. The first laser polishing crossing was executed across the fabrication direction 1×0 and the roughness was measured in the fabrication direction.

3.1 Initial selective laser melting surface analysis

The characterization of the initial surface was examined quantitatively with the perthometer. Therefore, the surfaces of eight vertical build samples were measured in the fabrication direction (FD) of the selective laser melting process and perpendicular to the fabrication direction (along the layers), Figure 7. The measured mean surface roughness Ra and the mean roughness depth Rz in both directions from eight initial samples are shown in Table 2. When considering Ra and Rz, the surface roughness is almost independent from the fabrication direction at those vertical samples. The spectral roughness by Fourier transformation, it becomes clear that the medium and long wavelength roughness structures larger than 40 μm represent the dominant part of the surface roughness, Figure 8. Between the orientations, only comparatively small differences in the spatial roughness greater than 156 μm can be observed. While the roughness in fabrication direction (FD) between 156 μm and 625 μm is greater than perpendicular to fabrication direction. Longer surface structures greater than 625 μm are lower, compared to perpendicular fabrication direction.

Figure 9a shows the 3D surface topography of the used parts, measured at the middle of the sample surface. The surface exhibits an orientation independent roughness structures and waviness, which is superimposed with particular protruding material accumulations. The micro structure, analysed by scanning electron microscopy shows the presence of oxides, which appear as white in colour (white areas) and residual particles, Figure 9b. Laser cleaning is used as a pre-process before laser polishing with the process parameters according chapter 2.1. It can remove these residual particles and oxides without creating a surface structuring or ablation of the base material, Figure 9c.

3.2 Pulsed mode polishing

3.2.1 Influence of focal position and laser beam intensity

In the pulsed wave mode, the focal position z was varied relative to the surface of the samples from +6 to +20 mm. Therefore, the beam diameter changes between 760 μm and 2020 μm, Figure 6b. The polishing study was done with a constant pulse energy of 0.51 J (laser power P_l = 1700 W, pulse duration  = 0.3 ms) in combination with an axis velocity of 50 mm/min. All fields were polished twice in the same direction (2×0). Further constant process parameters are shown in Table 1. The achievable roughness Ra and Rz is given over the focal position, Figure 10.

At a focal position of 6 mm, a roughness of Ra = 0.43 μm/Rz = 2.0 μm was achieved. An increase of the focal position and as a consequence thereof, a higher spot diameter and pulse overlap, the roughness was reduced to a minimum of Ra = 0.18 μm/Rz = 1.4 μm at a focal position of 12 mm. At this position, the laser beam had a spot diameter of 1298 μm and the pulse overlap was 84.6 %. A further rise of the focal position leads again to higher roughness. The scattering bars, which display the maximum deviation, also indicate that the focal position of 12 mm had a very high process stability.

Based on the achieved results, the focal position was set to 12 mm, Figure 10. In the second measuring series, the pulsed laser power was varied. Thereby, the laser pulse power was changed between 1200 W and 2000 W. Due to at a focus position of 12 mm, the laser beam mean intensity varies between 907 W/mm² and 1512 W/mm². In Figure 11, the influence of the laser power on the surface roughness is given.

A laser power of 1200 W/intensity of 907 W/mm² reduced the roughness to Ra = 0.43 μm, Rz = 5.96 μm. A rising laser peak power reduces stepwise the roughness to a minimum at a laser power of 1800 W, intensity of 1361 W/mm² to Ra = 0.15 μm, Rz = 1.33 μm. With a further increased laser power, the residual roughness is rising again. The surface roughness variation shows the same trend. With increasing surface quality the roughness variation gets significantly reduced. During the emission of a pulsed laser power of 1700 W, the average mean laser power is 510 W.

3.2.2 Resulting surface behaviour depending on the pulse overlap and laser beam intensity

Too high laser beam intensities in combination with a minor pulse overlap leads to a surface structuring of the single laser pulses, Figure 12a. At a laser beam intensity of 3571 W/mm² (z = 6 mm) and a pulse overlap of 73.7 % periodic pulse structures with a pulse distance in the range of 550 μm increases the surface roughness in the long wavelength range, Figure 12a. The micro structure of the polished surface is completely dense and free of porosity, Figure 12d. With decreasing laser beam intensity in combination with increasing pulse overlap, the single laser pulses become blurred, which leads to a decreasing surface roughness, Figure 12b. At a beam intensity between 1300 W/mm² and 1000 W/mm², where the lowest arithmetic roughness is achieved, the surface exhibits some open pores and linear structures in the dimension of 5 μm to 20 μm. Too low laser beam intensity, for example at a laser beam intensity I_avg of 531 W/mm² (z = 20 mm) results in a wavy and incomplete remelted surface, Figure 12c. The scanning electron microscopy image shows residual incompletely melted material accumulations or powder particles with dimensions up to 50 μm, Figure 12f.

3.2.3 Influence of axis velocity and machining strategy

A further important parameter is the track overlap. It affects the number of ‘remeltings’ and the energy input of an area by multiple exposures. Additionally, the track overlap, respectively the axis velocity has an impact on the resulting process speed, given by the area rate. In the following investigation, the axis velocity of the laser head was changed from 40 mm/min to 70 mm/min by steps of 10 mm/min. Thereby, the track overlap TO varies between 89.0 % and 93.7 %. Further, the introduced energy input ED, as a function of the laser power and the axis velocity, is varying between 43.7 J/mm² and 76.5 J/mm². The parameter combinations were applied to a single pass (1×0) and two passes (2×0). The polishing orientation was done perpendicular to fabrication direction, Figure 7. The achievable roughness is decreasing with a lower axis velocity v_f, increasing track overlap TO and energy density ED from Ra = 1.1 μm to Ra = 0.66 μm for a single pass, Figure 13. With a second crossing from the same direction (2×0), the roughness reduction is significantly increased in comparison to a single crossing at comparable velocity. The lowest Ra value of 0.31 μm is reached with  = 40 mm/min and two crossings (2×0).

If laser polishing is done twice with the same laser process parameters at crossed laser polishing direction (1×1), the achievable polishing quality is increased at an identical area rate in comparison to (2×0). Thereby the first crossing is done across the fabrication direction (identical to 1×0 and 2×0). The second crossing is reoriented by 90° degree perpendicular to the first crossing and thus polished in the fabrication direction. At 70 mm/min and one crossing in each direction (1×1), a Ra value of 0.21 μm is reached, Figure 13. In comparison, double polishing from the same direction (2×0) leads to Ra = 0.56 μm. With decreasing axis velocity the roughness is reduced to 0.15 μm Ra. When laser polishing is done with two crossings from each direction (2×2) only minimal improvements from 1×1 are possible to a minimum Ra value of 0.14 μm. The influence of the axis velocity, the track overlap and the energy density on the resulting polishing quality are decreasing with multiple crossings.

The laser polished fields were measured additionally across fabrication direction in order to analyse the homogeneity of the achieved surface, Figure 7 (measurement directions). In Figure 14, the influence of the number of passes and the machining strategy are depicted. The number of passes and the polishing direction have a major impact on the homogeneity of the surface. With single crossing the arithmetic roughness Ra can be twice as high perpendicular to the fabrication direction than parallel to the fabrication direction (40 mm/min (1×0)). With two crossings, the gap is greatly reduced. The best homogeneity is achieved by multiple polishing from different directions (1×1, 2×2). At four passes (2×2) the roughness becomes orientation independent. Additionally, the roughness scattering in the measured directions is decreasing significantly with further polishing passes.

A more detailed analysis of the remaining surface structures is given by the Fourier transformation of the measured surface profile. The residual surface structures depending on the number of passes and the polishing strategy at an axis velocity of 40 mm/min regarding to the main surface structures is shown in Figure 15. The micro roughness at structure wavelengths less than 31.25 μm is almost constant. In this range multiple polishing results in no significant improvement. Polishing with two passes from one direction (2×0) leads to a clear recognizable roughness improvement at the long structure wavelengths, especially at more than 125 μm. While single polishing exhibits still visible long roughness structures, after polishing twice the waviness is visibly reduced, Figure 16a, c. Polishing twice from crossed directions (1×1) reveals a further improvement in the range of 125 μm to 800 μm. In addition, the measured scatter of the surface quality decreases significantly. Four times laser polishing has only a very small improvement in the long surface structure range from 500 μm to 800 μm. The microscopic images of the surface appearance does not show any differences, Figure 16 b, d. The roughness improvement shown by multiple polishing at the long structure wavelengths is also depending on the need of a greater liquid material rearrangement, which cannot be reached within single laser pulses. The flattening of short-wave structures requires only a slight material rearrangement from the surface elevations to the surface depressions, which is reached within the created melt pool of one laser pulse. In contrast, smoothing of long surface structures needs a significantly greater liquid material rearrangement, which cannot be reached within one polishing pass.

3.3 Continuous mode polishing

3.3.1 Dependencies between beam intensity, track overlap and energy density

When polishing with a continuous laser, the pulse repetition rate is no longer a limiting factor by the laser source. So it is possible to work with higher pendulum frequencies and axis velocities. This offers the opportunity to work at higher average laser power. On the basis of pre investigations, a five times larger pendulum frequency and a focal position of 12 mm were found to be suitable parameters. In this study, the laser power was varied between 800 W and 1400 W, which leads in combination with a laser beam diameter of 1298 μm to a laser beam intensity from 604 W/mm² to 1058 W/mm². The axis feed rate was changed between 100 mm/min and 400 mm/min, which leads to a track overlap between 87.4 % and 96.8 %. Thereby the energy density ED varies between 12 J/mm² to 84 J/mm². The investigated process parameter combinations and the resulting energy density can be taken from Table 3. The investigations on laser polishing was done with a single crossing 1×0. Preliminary tests have shown, at axis velocities below 200 mm/min in combination with a laser power of 1400 W, a significantly rising melt pool depth and are not further investigated.

Table 3. Process parameter values for continuous mode laser polishing with the resulting laser beam intensity I_avg, track overlap TO and Energy density ED depending on the laser power P_l and axis velocity v_f.

Table 3. Prozessparameterwerte beim Laserpolieren im kontinuierlichen Betriebsmodus mit der resultierenden Laserstrahlintensität I_avg, der Spurüberlappung TO und der Energiedichte ED in Abhängigkeit von der Laserleistung P_l und der Achsgeschwindigkeit v_f .

Energy density ED [J/mm2]	Laser power Pl	800 W	1000 W	1200 W	1400 W
	Average beam intensity Iavg	604.2 W/mm2	755.3 W/mm2	906.3 W/mm2	1057.4 W/mm2
Axis velocity vf	100 mm/min	48.0	60.0	72.0	Not suitable
Track overlap TO	96.8 %
Axis velocity vf	150 mm/min	32.0	40.0	48.0	Not suitable
Track overlap TO	95.3 %
Axis velocity vf	200 mm/min	24.0	30.0	36.0	42.0
Track overlap TO	93.7 %
Axis velocity vf	250 mm/min	19.2	24.0	28.8	33.6
Track overlap TO	92.1 %
Axis velocity vf	300 mm/min	16.0	20.0	24.0	28.0
Track overlap TO	90.5 %
Axis velocity vf	350 mm/min	13.7	17.1	20.6	24.0
Track overlap TO	89.0 %
Axis velocity vf	400 mm/min	12.0	15.0	18.0	21.0
Track overlap TO	87.4 %

The roughness is shown as a function of axis velocity and track overlap at different laser powers, Figure 17. At 800 W laser power the roughness reduction is small. As the axis velocity increases, the roughness continues to rise from Ra 1.32 μm up to 3.0 μm. With increasing laser power up to 1400 W the achievable roughness decreases and the influence of the axis velocity gets lower and almost vanishes at 1400 W. The roughness in Ra is varying between 0.22 μm and 0.28 μm. The absolute minimum surface roughness is reached at an axis velocity of 200 mm/min.

The energy density, used for continous wave laser polishing, is a function of the laser power P_l and the axis velocity v_f. Figure 18 shows the same results as in Figure 17, however, as a function of energy density. At low laser beam intensities, especially at a laser power of 800 W the roughness decreases with increasing energy density. At higher beam intensities, the achievable surface quality becomes independent from the introduced energy per area. Further, it can be seen that the main process parameter, which has a great influence on the arithmetic surface roughness Ra, is the laser beam intensity.

3.3.2 Roughness improvement by multiple polishing passes

According to the investigations on pulsed mode laser polishing, CW polishing was also done with multiple polishing passes. Therefore the parameters with the lowest roughness from section 3.4 (P_L = 1400 W, v_f = 250 mm/min) were used, Figure 19. Similar to polishing at the pulsed mode, CW polishing with one crossing exhibited a significant influence of the measuring direction on the roughness. With a second pass from the same direction (2×0) the roughness decreases to Ra = 0.14 μm (in fabrication direction), respectively 0.17 μm (perpendicular to fabrication direction). Polishing from both directions (1×1) leads to a similar roughness value. Here the roughness perpendicular to fabrication direction is lower than the roughness in fabrication direction. Polishing twice from both directions (2×2) revealed an almost homogeneous average roughness Ra in the range of 0.14 μm–0.15 μm at a high repeatability.

When considering the residual roughness after laser polishing over the spatial wavelength an additional pass from the same direction (2×0) leads to a roughness reduction over the complete roughness spectrum, Figure 20. In contrast to the pulsed mode polishing the crossed polishing strategy with two or four passes (1×1, 2×2) reveals no further significant improvement, especially at long roughness structures.

The microscopic surface at one polishing pass exhibits linear structures and several pores in the dimension of 2 μm–10 μm, Figure 21a. With 2×2 polishing passes those structures can be almost eliminated. Remaining submicron structures on an otherwise flat surface are still visible, Figure 21b.

3.4 Comparison between the laser operation modes

Laser polishing with pulsed wave (PW) and continuous wave (CW) laser radiation was investigated. Table 4 displays the differences of the main process parameters between both laser modes with the highest roughness reduction and the achieved surface roughness with one polishing pass and multiple polishing passes. The continuous wave-polishing process is operating at higher average laser power of 1400 W in comparison to 510 W at pulsed wave. The laser beam intensity at continuous wave between 907 W/mm² and 1058 W/mm² is significantly lower compared to the pulsed wave-mode with 1300 W/mm², while the track overlap at the ‘sweet spot’ is almost the same. The main differences between both operation modes is the achievable area rate, which is with continuous wave mode polishing five times higher. As a cause of the significant slower process in the pulsed wave mode as a consequence of the pulse frequency limitation and resulting increased thermal losses from the process zone, the energy density of 76.5 J/mm² is significantly higher in comparison to the continuous wave-polishing (42.0 J/mm²).

With a single laser polishing pass the lowest achievable roughness is by continuous wave mode polishing being almost one third in comparison to pulsed wave mode laser polishing. A more detailed analysis of the remaining surface structures is given by the Fourier transformation of the measured surface profile. Figure 22 shows the differences between the pulsed and continuous laser operation mode with one polishing pass and the optimized parameters from Table 4.

At surface structures with small structure wavelengths in the range of 2.5 μm–7.8 μm, both laser operation modes lead to a similar residual roughness. At the medium spatial wavelength between 7.8 μm and 125 μm pulsed mode laser polishing reaches a higher roughness reduction. The long structure wavelengths up to 800 μm can be better reduced at one polishing pass by continuous laser radiation. Flattening of micro roughness with small structure wavelengths causes a very minor rearrangement of liquid material. However, flattening of long wavelength structures affords a widespread melt pool, where liquid material can be rearranged. The continuous energy input at continuous wave laser polishing in combination with a more than five times larger beam velocity may lead to a larger continuous melt pool, compared to pulsed wave mode laser polishing, which seems to be better for large surface structure wavelengths.

When comparing the 3D topography of the polished surfaces, measured by the profilometer, further differences can be recognized. Single laser polishing at the continuous laser mode exhibits a surface unevenness with distances above the tactile measureable spatial wavelength, Figure 23a. The border of the polishing field at the start of polishing and at the turning points of the scanner beam deflection is at the same high as the surrounding initial surface. As the polishing process progresses in the direction of the axis velocity v_f, the surface in the outer areas rises by rearranging of molten material. At the end of polishing along the pendulum movement of the scanner an indentation occurs, also called end crater, visible by the height profile with a varying depth of 10 μm–20 μm, Figure 23a. In comparison, pulsed laser polishing has no ‘ditch’ at the end of polishing. However, pulse lenses with superimposed waviness are visible. At the turning points of the scanner movement much material is deposited, which leads to a surface super elevation of up to 30 μm, Figure 23b.